Real-world outcomes of second-line therapy in advanced thymic carcinoma: a multicenter Turkish Oncology Group (TOG) study

Abstract

Thymic carcinoma (TC) is a rare and aggressive malignancy with limited evidence guiding second-line treatment after prior systemic therapy. We evaluated real-world outcomes of second-line systemic therapy in a multicenter cohort. Adults with unresectable, recurrent, or metastatic TC who received second-line systemic therapy after prior systemic therapy between 2010 and 2023 were retrospectively analyzed. Treatments were categorized as multidrug chemotherapy, gemcitabine monotherapy, PD-1 inhibitor therapy, or targeted therapy (sunitinib). For the primary comparison of multidrug chemotherapy versus gemcitabine monotherapy, propensity scores based on age, sex, ECOG performance status, Masaoka–Koga stage IV, prior surgery, and calendar year were used to generate overlap weights. Weighted Cox models used robust standard errors clustered by treatment center. A sensitivity propensity-score model additionally included first-line carboplatin–paclitaxel exposure (yes/no). A total of 107 patients from 20 treatment centers were included. Median age was 53.2 years (IQR, 41.5–58.0), 78 (72.9%) were male, 81 of 106 patients with available data (76.4%) had ECOG 0–1, and 61 (57.0%) had Masaoka–Koga stage IV disease at diagnosis. At a median follow-up of 35 months, median PFS and OS for the overall cohort were 13.6 and 35.0 months, respectively. In the primary overlap-weighted comparison ( n  = 89), multidrug chemotherapy was associated with a lower hazard of progression or death than gemcitabine monotherapy (HR, 0.29; 95% CI, 0.14–0.59; p  < 0.001) and a lower hazard of death (HR, 0.42; 95% CI, 0.18–0.97; p  = 0.043). In a sensitivity model additionally incorporating first-line carboplatin–paclitaxel exposure, the PFS estimate remained significant (HR, 0.32; 95% CI, 0.16–0.65; p  = 0.0015), whereas the OS estimate remained directionally similar but was less precise (HR, 0.46; 95% CI, 0.19–1.11; p  = 0.083). The PD-1 inhibitor group ( n  = 13) had median PFS and OS of 15.0 and 48.0 months, respectively, in unadjusted descriptive analyses. In this retrospective cohort, multidrug chemotherapy was associated with lower hazards of progression or death and of death than gemcitabine monotherapy in the primary overlap-weighted analysis. A sensitivity analysis additionally accounting for first-line carboplatin–paclitaxel exposure preserved the PFS association and yielded a similar but less precise OS estimate. Outcomes observed with PD-1 inhibitors were favorable but should be interpreted descriptively because of the small sample size, later treatment era, and potential residual confounding. These findings are hypothesis-generating and require prospective validation.